INDEX

Explanations

words related to the physical environment

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "environment."

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

sonian

-0.90

headed

-0.90

doms

-0.85

mad

-0.84

heading

-0.82

ief

-0.80

uster

-0.80

thood

-0.79

leaf

-0.79

ilts

-0.79

POSITIVE LOGITS

 environment

0.95

 environments

0.92

 Environment

0.90

 variables

0.84

Variable

0.80

 conducive

0.79

 variable

0.77

eers

0.74

 atmosp

0.72

 conditions

0.71

Activations Density 0.020%

words related to the physical environment

references to the concept of "environment."

No Comments

No Known Activations