INDEX

Explanations

words related to a positive or desirable environment or situation

oai_token-act-pair · gpt-3.5-turbo

references to environments or ecological contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 actionGroup

-0.78

butt

-0.78

Reviewer

-0.77

Ô

-0.72

quartered

-0.71

head

-0.68

esome

-0.68

soever

-0.65

dress

-0.64

drive

-0.64

POSITIVE LOGITS

env

1.10

isions

1.09

ENTION

0.89

env

0.88

ours

0.83

olve

0.82

ision

0.79

ounter

0.79

utions

0.75

igrated

0.75

Activations Density 0.015%

words related to a positive or desirable environment or situation

references to environments or ecological contexts

No Comments

No Known Activations