INDEX

Explanations

steps or instructions in a sequence

oai_token-act-pair · gpt-3.5-turbo

step-by-step instructions or sequences

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

luaj

-0.76

ãĥĩãĤ£

-0.69

ãĥ´

-0.69

ITED

-0.66

uci

-0.64

tumblr

-0.63

ãĤª

-0.63

United

-0.61

cov

-0.61

ãĥĥ

-0.61

POSITIVE LOGITS

 steps

0.75

udic

0.75

 Steps

0.72

steps

0.71

rity

0.71

ĵĺ

0.71

acter

0.70

 paddle

0.68

Goal

0.67

QUIRE

0.67

Activations Density 0.100%

steps or instructions in a sequence

step-by-step instructions or sequences

No Comments

No Known Activations