INDEX

Explanations

phrases related to job terminations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "laid" used in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

externalActionCode

-0.76

=-=-=-=-=-=-=-=-

-0.75

ively

-0.73

ities

-0.73

EVA

-0.72

inho

-0.69

arta

-0.68

acca

-0.67

ernaut

-0.65

Guard

-0.65

POSITIVE LOGITS

 laid

0.91

sheets

0.82

ndra

0.82

\\\\\\\\

0.82

galitarian

0.79

 showc

0.77

 groundwork

0.76

tons

0.76

ayer

0.75

bare

0.75

Activations Density 0.010%

phrases related to job terminations

instances of the word "laid" used in various contexts

No Comments

No Known Activations