INDEX

Explanations

action verbs related to advice or recommendations

oai_token-act-pair · gpt-3.5-turbo

phrases that express recommendations or concerns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 ..............

-0.69

 ........

-0.69

 ................

-0.64

'.

-0.63

ryu

-0.59

anners

-0.59

.........

-0.58

kamp

-0.57

liv

-0.57

."

-0.57

POSITIVE LOGITS

ever

0.66

active

0.63

 improvement

0.63

 relates

0.62

 involves

0.61

 bothers

0.61

 pursu

0.60

 occurs

0.59

 varies

0.59

%]

0.58

Activations Density 0.374%

action verbs related to advice or recommendations

phrases that express recommendations or concerns

No Comments

No Known Activations