INDEX

Explanations

words related to items or concepts that the subject is handling or interacting with

oai_token-act-pair · gpt-3.5-turbo

references to specific items or categories indicated by demonstrative pronouns and determiners

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

OSP

-0.69

RT

-0.69

Ax

-0.67

seek

-0.66

managed

-0.66

Supported

-0.65

amia

-0.64

alter

-0.63

rontal

-0.63

ARB

-0.63

POSITIVE LOGITS

 leftover

1.07

 unused

0.96

 excess

0.91

 entire

0.89

 offending

0.89

 remaining

0.84

 envelope

0.79

 belongings

0.78

 aside

0.74

 onto

0.73

Activations Density 0.287%

words related to items or concepts that the subject is handling or interacting with

references to specific items or categories indicated by demonstrative pronouns and determiners

No Comments

No Known Activations